[litv] Support 'promo' URLs (closes #10385)

[youtube-dl] / youtube_dl / extractor / common.py

diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py
index 67f49f51b52c85c488851903e309c38c217f23dc..07d58afe7bb2a89a70a61dba447723e1518cc873 100644 (file)
--- a/youtube_dl/extractor/common.py
+++ b/youtube_dl/extractor/common.py
@@ -662,6 +662,24 @@ class InfoExtractor(object):
         else:
             return res
 
+    def _get_netrc_login_info(self, netrc_machine=None):
+        username = None
+        password = None
+        netrc_machine = netrc_machine or self._NETRC_MACHINE
+
+        if self._downloader.params.get('usenetrc', False):
+            try:
+                info = netrc.netrc().authenticators(netrc_machine)
+                if info is not None:
+                    username = info[0]
+                    password = info[2]
+                else:
+                    raise netrc.NetrcParseError('No authenticators for %s' % netrc_machine)
+            except (IOError, netrc.NetrcParseError) as err:
+                self._downloader.report_warning('parsing .netrc: %s' % error_to_compat_str(err))
+
+        return (username, password)
+
     def _get_login_info(self):
         """
         Get the login info as (username, password)
@@ -679,16 +697,8 @@ class InfoExtractor(object):
         if downloader_params.get('username') is not None:
             username = downloader_params['username']
             password = downloader_params['password']
-        elif downloader_params.get('usenetrc', False):
-            try:
-                info = netrc.netrc().authenticators(self._NETRC_MACHINE)
-                if info is not None:
-                    username = info[0]
-                    password = info[2]
-                else:
-                    raise netrc.NetrcParseError('No authenticators for %s' % self._NETRC_MACHINE)
-            except (IOError, netrc.NetrcParseError) as err:
-                self._downloader.report_warning('parsing .netrc: %s' % error_to_compat_str(err))
+        else:
+            username, password = self._get_netrc_login_info()
 
         return (username, password)
 
@@ -816,11 +826,14 @@ class InfoExtractor(object):
         json_ld = self._search_regex(
             r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>',
             html, 'JSON-LD', group='json_ld', **kwargs)
+        default = kwargs.get('default', NO_DEFAULT)
         if not json_ld:
-            return {}
-        return self._json_ld(
-            json_ld, video_id, fatal=kwargs.get('fatal', True),
-            expected_type=expected_type)
+            return default if default is not NO_DEFAULT else {}
+        # JSON-LD may be malformed and thus `fatal` should be respected.
+        # At the same time `default` may be passed that assumes `fatal=False`
+        # for _search_regex. Let's simulate the same behavior here as well.
+        fatal = kwargs.get('fatal', True) if default == NO_DEFAULT else False
+        return self._json_ld(json_ld, video_id, fatal=fatal, expected_type=expected_type)
 
     def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None):
         if isinstance(json_ld, compat_str):
@@ -1140,7 +1153,7 @@ class InfoExtractor(object):
             'url': m3u8_url,
             'ext': ext,
             'protocol': 'm3u8',
-            'preference': -100,
+            'preference': preference - 100 if preference else -100,
             'resolution': 'multiple',
             'format_note': 'Quality selection URL',
         }
@@ -1682,7 +1695,7 @@ class InfoExtractor(object):
                         self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
         return formats
 
-    def _parse_html5_media_entries(self, base_url, webpage):
+    def _parse_html5_media_entries(self, base_url, webpage, video_id, m3u8_id=None):
         def absolute_url(video_url):
             return compat_urlparse.urljoin(base_url, video_url)
 
@@ -1697,6 +1710,21 @@ class InfoExtractor(object):
                 return f
             return {}
 
+        def _media_formats(src, cur_media_type):
+            full_url = absolute_url(src)
+            if determine_ext(full_url) == 'm3u8':
+                is_plain_url = False
+                formats = self._extract_m3u8_formats(
+                    full_url, video_id, ext='mp4', entry_protocol='m3u8_native',
+                    m3u8_id=m3u8_id)
+            else:
+                is_plain_url = True
+                formats = [{
+                    'url': full_url,
+                    'vcodec': 'none' if cur_media_type == 'audio' else None,
+                }]
+            return is_plain_url, formats
+
         entries = []
         for media_tag, media_type, media_content in re.findall(r'(?s)(<(?P<tag>video|audio)[^>]*>)(.*?)</(?P=tag)>', webpage):
             media_info = {
@@ -1706,10 +1734,8 @@ class InfoExtractor(object):
             media_attributes = extract_attributes(media_tag)
             src = media_attributes.get('src')
             if src:
-                media_info['formats'].append({
-                    'url': absolute_url(src),
-                    'vcodec': 'none' if media_type == 'audio' else None,
-                })
+                _, formats = _media_formats(src)
+                media_info['formats'].extend(formats)
             media_info['thumbnail'] = media_attributes.get('poster')
             if media_content:
                 for source_tag in re.findall(r'<source[^>]+>', media_content):
@@ -1717,12 +1743,13 @@ class InfoExtractor(object):
                     src = source_attributes.get('src')
                     if not src:
                         continue
-                    f = parse_content_type(source_attributes.get('type'))
-                    f.update({
-                        'url': absolute_url(src),
-                        'vcodec': 'none' if media_type == 'audio' else None,
-                    })
-                    media_info['formats'].append(f)
+                    is_plain_url, formats = _media_formats(src, media_type)
+                    if is_plain_url:
+                        f = parse_content_type(source_attributes.get('type'))
+                        f.update(formats[0])
+                        media_info['formats'].append(f)
+                    else:
+                        media_info['formats'].extend(formats)
                 for track_tag in re.findall(r'<track[^>]+>', media_content):
                     track_attributes = extract_attributes(track_tag)
                     kind = track_attributes.get('kind')